Critical Trends Assessment Program Monitoring Protocols

Illinois Natural History Survey Office of the Chief

Technical Report 2002-2

CTAP Monitoring Protocols From the Editor:

The Critical Trends Assessment Program (CTAP) is a long-term endeavor, which monitors the condition of forests, wetlands, grasslands, and streams throughout the state of Illinois. It assesses current and future trends in ecological condition on statewide, regional, and site-specific bases. This program, an endeavor of the Illinois Department of Natural Resources, is unique because it is the first-ever attempt at a statewide comprehensive assessment undertaken by a state natural resource organization. A total of 600 sites representing four habitats (150 of each; 30 sites per habitat per year) were randomly selected from across the state on both public and private land. Since 1997 the CTAP professional scientists of the Illinois Natural History Survey (INHS) have been conducting surveys at these sites.

In this document we present the CTAP standardized protocols for monitoring forests,

wetlands, grasslands, and streams. In forests, wetlands, and grasslands data on herbaceous and woody vegetation, birds, and insects are collected. In streams, aquatic insects (EPT taxa: Ephemeroptera [mayflies], Plecoptera [stoneflies], and Trichoptera [caddisflies]) are the primary assemblage used. Each main section (terrestrial or aquatic) can stand as an independent document. This explains, in part, differences in format and repetition of information. The latter is more obvious in the terrestrial monitoring protocols. All groups, organizations, and individuals are welcome and encouraged to use these monitoring protocols. Following these protocols will benefit the user in several ways. For example, by collecting similar data, the user will be able to incorporate our data into her/his project. For additional information about our program and data, go to the following web page: http://ctap.inhs.uiuc.edu. Finally, this document should be cited as follows: For general reference to the document: Molano-Flores, B. 2002. Critical Trends Assessment Program Monitoring Protocols. Illinois Natural History Survey, Office of the Chief, Technical Report 2002-2, Champaign, IL. 38 pp, + Figures, Tables, and Appendix. or For each section of the document: Author(s) section. 2002. Section title. in B. Molano-Flores (ed.) Critical Trends Assessment Program Monitoring Protocols. Illinois Natural History Survey, Office of the Chief Technical Report 2002-2, Champaign, IL. 38 pp, + Figures, Tables, and Appendix.

CTAP Protocols i

Document Contributors: Several people have written, provided input, and reviewed this document. Below we

provide a list of the people involved in the production of this document. To all of them thank you. Document Editor: Brenda Molano-Flores Section Authors:

Statistical Sampling Design: Jeff Brawn, Liane Sulaway, Ellen Brewer, Dan Niven, Chris Phillips, and William Ruesink Initial study site selection: Mark Joselyn, Ellen Brewer, and Liane Sulaway Habitat criteria for study sites: John Taft, Ken Robertson, Scott Robinson, Jeff Brawn, Chris Phillips, Dan Niven, R. Edward DeWalt, and Larry Page Site evaluation, site selection, and documentation: Rhetta Jack, Steve Bailey, Connie Carroll, and Cynthia Dassler

Plant Sampling Protocols: Connie Carroll, Cynthia Dassler, James Ellis, Greg Spyreas, John Taft, and Kenneth Robertson

Bird Sampling Protocols: Dan Niven, Steve Bailey, Rhetta Jack, Jeff Brawn, and Scott Robinson

Terrestrial Insect Sampling Protocols: Chris Dietrich, Michelle Biyal, and Sue Gallo

Aquatic monitoring protocols and Appendix: R. Edward DeWalt

CTAP Protocols ii

Table of Contents

Page No. I- Terrestrial monitoring protocols 1 1) Statistical Sampling Design 1 a) Sampling unit 1 b) Township selection 1 c) Sample sizes 1 2) Initial study site selection 2 a) Forest 2 b) Wetland 2 c) Grassland 3 3) Habitat criteria for study sites 3 a) Forest habitat criteria 3 b) Wetland habitat criteria 4 c) Grassland habitat criteria 5 4) Site evaluation, site selection, and documentation 6 a) Landowner access 7 b) Site evaluations, ground truthing 8 c) Establishing a study plot 9 d) Documentation 9 5) Plant Sampling Protocols 11 a) Forest Sampling Protocols 11

i) Establishing study plots 11 ii) GPS data 11 iii) General site characteristics 11 iv) Slope and aspect 12 v) Photographs 12 vi) Ground cover (including woody cover < 1m tall) 12 vii) Woody vegetation < 5cm dbh, but at least 1m 12 viii) Woody vegetation > 5cm dbh 13 ix) Big Plot 13 x) Collection of voucher specimens 13

CTAP Protocols iii

b) Wetland Sampling Protocols 14 i) Establishing study plots 14 ii) GPS data 14 iii) General site characteristics 15 iv) Slope and aspect 15 v) Photographs 15 vi) Ground cover (including woody cover < 1m tall) 15 vii) Woody vegetation < 5cm dbh, but at least 1m 15 viii) Woody vegetation > 5cm dbh 15 ix) Big Plot 16 x) Collection of voucher specimens 16

c) Grassland Sampling Protocols 17

i) Establishing study plots 17 ii) GPS data 17 iii) General site characteristics 17 iv) Slope and aspect 18 v) Photographs 18 vi) Ground cover (including woody cover < 1m tall) 18 vii) Woody vegetation < 5cm dbh, but at least 1m 18 viii) Woody vegetation > 5cm dbh 18 ix) Big Plot 18 x) Collection of voucher specimens 19

6) Terrestrial Insect Sampling Protocols 20 a) Sampling locations 20 b) Sampling methods 20 c) Notes on sampling methods 22 d) Collection numbers 22

7) Bird Sampling Protocols 23

a) Requirements and yearly preparation 23 b) Establishing census points 23 c) Recording habitat data 24 d) When to census 24 e) Acceptable weather 25 f) Conducting point counts 25 g) Playback tapes in wetlands 26 h) Additional grasslands and wetlands for bird monitoring 26

CTAP Protocols iv

CTAP Protocols v

II - Aquatic monitoring protocols 27 1) Selection of Random Stream Locations 27 2) Acceptability of Stream Segments 27 3) Establishing a Sample Reach 27 4) Geographic Referencing of the Sample Reach 28 5) Photographs 28 6) Aquatic Insect Sampling Methods 28

a) Phenology of Sampling 28 b) Sampling Protocols 29

i) High-Energy Habitats 29 ii) Low-Energy Habitats 29

c) Habitat Quality Assessment 29 d) Water Chemistry 30 e) Sample Processing, Vouchering of Specimens 30

III - Literature Cited 31 IV - Tables and Figures 33 a) Terrestrial monitoring protocols 33 i) Figures 33 ii) Tables 34 b) Aquatic monitoring protocols 34

i) Figures 34

V – Appendix 35 Logistics for using EPT to indicate stream health 35

I- Terrestrial monitoring protocols 1) Statistical Sampling Design a) Sampling unit In order to draw statistical inference about the status and trends of bio-indicators at the statewide level, a population of random, independent sampling units is needed. The population of 1765 Illinois Public Land Survey townships was chosen as the sampling units. The average size of these townships is 20,420 acres (SD = 6,612, range = 29,632). With the exception of townships along the edge of the state, along rivers, or along meridians, most townships are approximately square in shape and 6 miles on a side, and composed of 36 1-mile2 “sections” (a township of this size would be 23,040 acres). Townships were chosen as the sampling unit because: 1) they are large enough to assure that suitable habitats for sampling will be found in most randomly selected townships; 2) there are enough townships to sample from; and 3) their common usage throughout the State makes them a convenient unit for GIS scientists to work with. Other grid systems were considered but rejected. For example, the EMAP hexagonal grid system used for some federal monitoring programs (such as the U.S. Forest Service’s “Forest Health Monitoring” program) was rejected because the population of hexagonal polygons in the State was too small. b) Township selection For each habitat type to be monitored (forest, grassland, and wetland), the state’s 1765 townships were randomly ranked. Ranking the townships during randomization allows us to avoid the bias of subjectively choosing which randomly selected townships to sample in a given year. Sampling proceeds sequentially down the list until an appropriate number of townships have been sampled (see below). In order to make inferences about statewide environmental conditions, every location in the State should have an equal probability of being sampled. Therefore, to avoid bias toward sampling areas in larger townships, each township was weighted by its area during the randomization procedure. Fig. 1a-c shows the location of the first 50 randomly selected townships. They are ranked from 1-50 and also include the unique township number. The goal is to monitor the townships ranked 1-30 in the first year, 31-60 the second year, etc. However, each year a few of the townships may have to be rejected because they do not have suitable habitat, reasonable access sites, and/or it is not possible to get permission to sample at the site. If this is the case, the scientists continue sequentially down the list until 30 acceptable townships are sampled. c) Sample sizes Statewide sampling will proceed on a five-year cycle. At least 30 new sites (townships) will be sampled in each of the four focal habitats every year for five years (i.e. sample without replacement), resulting in a total sample of approximately 150 townships per habitat. Although there will usually be more than one suitable location for sampling within each township, for any given habitat only one site will be sampled in each township. This will be done even though there may be multiple sampling locations (e.g. transects) at each site. The sole exception to the one site sampling per township will be made with bird sites in wetlands and grasslands. Additional sites may be monitored in a township (see Bird Sampling Protocols).

CTAP Protocols 1

Time and resource constraints and the results of analyses of statistical power will ultimately determine the exact number of townships to be sampled. After the first five years the sampling cycle will be repeated: in the sixth year townships visited in year one (i.e., 1997) will be revisited, with the possible addition of some new townships. These townships may be needed to replace sites that are no longer suitable and/or are no longer accessible. In addition, without the addition of new sites the monitoring program may only detect changes due to natural successional processes rather than continually gaining new input into the general conditions of natural resources across the State. 2) Initial study site selection Before CTAP biologists begin their field work each year, potentially suitable habitat for sampling locations must be identified from within each randomly selected township. This section describes the methods used to objectively identify these habitats and select potential study sites. Detailed GIS field maps are produced for each township to be evaluated and sampled. These maps show the distribution of land cover types and the location of potential study sites. Although multiple sites may be suitable in each township, only one location is sampled. This location is the lowest ranked acceptable site on which permission to sample is granted by the landowner. The sole exception to the one site sampling per township will be made with bird sites in wetlands and grasslands. Additional sites may be monitored in a township (see Bird Sampling Protocols). a) Forest The CTAP Land Cover of Illinois database (Illinois Department of Natural Resources, 1995) categorizes each pixel of land (approximately 90’ x 90’) into one of 19 land cover categories. Five of these land cover types describe forested land: deciduous closed canopy, deciduous open canopy, coniferous, forested wetland, and swamp. Together, these land cover types account for more than 13% of Illinois’ land cover. For the purpose of this monitoring project, the five “forest” categories have been pooled. For each randomly selected township, all pixels with forest cover that meet both of the following criteria were identified: 1) pixels that were part of a forest patch that was at least 20 acres in size; 2) pixels that were surrounded by a forest buffer of 114m (4 pixels). Within this available population of forest, the coordinates of a maximum of 50 points were randomly identified as potential monitoring sites in each township. These points were then randomly ranked (1-20) to provide a non-subjective order to follow when evaluating potential study sites. The land cover maps used in the field (Fig. 2) show the location of these 20 points. b) Wetland Potential sampling locations for wetlands were determined using the digital Illinois Wetlands Inventory database (Illinois Natural Resources Geospatial Data Clearinghouse, 1997; Suloway and Hubbell, 1994). The data for Illinois were generated from high altitude aerial photography acquired from 1980 – 1987; most of the photography was taken in 1983. This database may miss up to 40% of the State's wetlands, however, it is not biased to missing particular types of wetlands - it is only biased by size and "wetness"(Alan Plocher, Illinois Natural History Survey, pers. com.).

CTAP Protocols 2

Criteria used to identify potential wetland sampling sites were based on wetland type and size. Specifically, wetlands suitable for CTAP monitoring are dominated by emergent palustrine vegetation (i.e. rooted herbaceous hydrophytic vegetation such as sedges, rushes, forbs, and grasses) and they are greater than two acres in size. There were 16,542 discreet emergent wetlands larger than two acres known from within the State, totaling 166,256 acres (0.5% of Illinois), with a mean size of 10.1 acres. These emergent wetlands were randomly ranked (1-indeterminent) within each selected township to establish sampling priority and field maps have been produced which show their location (Fig. 3). c) Grassland Identification of sampling locations for grasslands was based on the Illinois Land Cover database. Two land cover classes, rural grassland and urban grassland, were used. Together they cover more than 19% of Illinois' land surface. The rural grassland category includes pastures, hayfields, idle fields, and non-agricultural land such as reclaimed mine land, road and railroad right-of-ways and remnant prairies. Urban grassland includes open space, parks and golf courses in urban areas. High quality grasslands (native prairie remnants) are rare in Illinois, and they are often very small. Because we did not want to exclude the possibility of sampling these sites, no size constraints were placed on patches of grassland selected for sampling. Specific locations for sampling were determined by randomly placing 50 ranked points within each selected township in areas classified as grassland in the Land Cover database. Field maps show the location of these 50 points (Fig. 4). 3) Habitat criteria for study sites Criteria have been established to objectively accept or reject sites after ground truthing because the habitat categories recognized by the land cover database are broad (e.g. open woodland may include city parks or relatively young successional woodland, as well as native savannas) and errors may have occurred in the classification of satellite images. In this way, monitoring is restricted to sites that are representative of the intended habitat type. Moreover, by discarding sample plots in highly divergent habitat types (such as pine plantations and city parks), undesirable variation between sites is reduced, which should provide higher statistical power to detect trends. The primary criteria for acceptance is that all sites be minimally to moderately managed, currently in a somewhat natural state and undergoing successional processes such that changes in condition will be possible and detectable. Potential monitoring sites selected by GIS (described in the previous section) are ground-truthed to determine if they meet the following criteria necessary for inclusion in the pool of sites to be sampled. A criterion common to all habitat types is that the area sampled by the field crew must be fully within the township being monitored. If the site is only of sufficient size to monitor if part of the sampling is conducted across the border of an adjacent township, then the site is rejected. a) Forest habitat criteria All five land cover types identified as “forest” in the Illinois Land Cover database are included in the pool of potentially acceptable monitoring sites for the purpose of determining if a plot meets the size criteria mentioned above (minimum acreage). Although this broad range of forest types

CTAP Protocols 3

may be suitable for bird monitoring, not all of these sites are acceptable for monitoring plants and insects. Thus, more restrictive criteria are necessary for sites to be acceptable for monitoring. Forest sites acceptable for CTAP monitoring meet the following criteria when assessed: • Sites have a diameter [radius] of at least 150m [75] of suitable homogeneous forest habitat.

The potential forest types, as categorized by CTAP, are moist/wet uplands (mesic to wet uplands and north-facing slopes), dry uplands (dry to dry-mesic uplands and south-facing slopes), and bottomlands. The site is big enough to include transects which are broken to accommodate crossing streams, trails, etc.

• Forest tracts average 75% canopy cover, although some areas within the tract may be more open due to selective logging or tree fall gaps. Not more than the equivalent of one transect falls within areas with less than 75% canopy cover.

• The majority of the trees in the forest tract are at least 10cm dbh. Exception: stunted “pygmy” woodland found on naturally xeric sites

• Forests currently lightly grazed are acceptable (unless the ground cover has been replaced by plantings of pasture grass or a manicured lawn).

• Sites marked to be logged or developed are acceptable as long as monitoring can be completed during the current field season.

The following forest sites are unacceptable: • Forests grazed to completely denuded of ground cover vegetation. • Sites that have extreme anthropogenic degradation factors such as ground cover replaced by

plantings of a pasture grass or a manicured lawn (e.g. forested city parks). • Plantations, unless the majority of the trees growing naturally beneath the ones planted are >

10cm dbh. • If during the year, when a forest is assessed, the water is to deep to safely work in, then it is

rechecked during the site assessment season. If the water is still too high at the end of the site assessment season, then the site is rejected.

• No access due to safety reasons or equipment (i.e., boats) Bird monitoring in forests occurs on a much larger spatial scale than plant monitoring (bird census points are spaced at least 150m apart – see below). Therefore, bird census locations are not restricted to a homogeneous forest type, but otherwise meet all the criteria mentioned above. b) Wetland habitat criteria The pool of potential, random monitoring sites was identified from the Illinois Wetlands Inventory database (IWI). A 2-acre IWI size minimum was used for the potential pool of random monitoring sites (based on logistic considerations) in the hopes that it would reduce the number of unacceptable sites that would need to be evaluated after ground-truthing.

CTAP Protocols 4

Wetland sites acceptable for CTAP monitoring meet the following criteria when assessed: • The minimum area of suitable habitat is 500m2 with a minimum width of 10m (e.g. 50m x

10m or the equivalent). • Sites have < 50% woody shrub or tree cover. • An area is considered a wetland if > 50% of the relative cover of dominant plant species are

wetland plants in the following categories: obligate, facultative wetland, or facultative (as defined in Admiraal et al. 1997).

• If open water is present, then the wetland is suitable if there is > 30% plant cover. • Artificially constructed wetlands and lightly grazed wetlands are suitable. • Sites scheduled to be plowed, drained, or developed are acceptable as long as monitoring can

be completed during the current field season. Sites with the following characteristics are not acceptable: • Ponds were excluded if the amount of emergent vegetation does not meet the criteria above. • If the wetland has been recently plowed (the year of the census), if it is currently being filled,

or if the wetland is unsafe to work in (i.e., water greater than 1m deep or too mucky to be safe), then it is discarded.

• If during the year, when a wetland is assessed, the water is to deep to safely work in, then it is rechecked during the site assessment season. If the water is still too high at the end of the site assessment season, then the site is rejected.

c) Grassland habitat criteria Native grasslands are currently almost nonexistent in Illinois. The once vast prairies have been almost totally replaced by agriculture or urban landscapes. However, “grassland” habitat, as characterized by the Illinois Land Cover database, still occupies 19 percent of the State's land cover (Illinois Department of Natural Resources 1996). Grasslands identified by the Land Cover of Illinois database include a diversity of habitat types such as pastures, hayfields, airfields, parks, cemeteries, abandoned fields, grassland strips along roadsides, and native prairie remnants. Most of these areas that are physiognomically classified as grasslands, have been planted or are heavily managed in other ways and are now dominated by the presence of exotic species. However, even though these disturbed habitats no longer have a long history of natural succession, disturbed sites may still harbor some native species that once occurred in prairies, and for some native species these disturbed grasslands may be the only refugia standing between them and local extinction. For these reasons CTAP biologists are monitoring a broad spectrum of grassland habitats. The primary criteria for accepting a grassland site for inclusion in the CTAP monitoring program is that the site be currently managed at a relatively low intensity. Grassland sites acceptable for CTAP monitoring meet the following criteria when assessed: • The minimum area of suitable habitat is 500m2, with a minimum width of 10m (e.g. 50m x

10m, or the equivalent).

CTAP Protocols 5

• The grasslands have < 50% shrub cover and < 50% canopy cover. • Sites scheduled to be plowed or developed are acceptable as long as monitoring can be

completed during the current field season. Examples of acceptable grassland sites include: • Ungrazed, abandoned, or lightly grazed pastures • Grasslands that have not recently been planted in monocultures • Areas planted in alfalfa or clover, if there is > 50% cover of other plant species present (if %

of other species is <50%, then it is considered a monoculture) • Abandoned agricultural fields, as long as they do not still contain crop stubble • Overgrown or infrequently mowed rights-of-way • Native prairies (including old cemeteries) • Old fields • Prairie reconstructions • CRP lands Unsuitable grassland sites include: • Fields or pastures that are heavily grazed (if not sure if it is heavily grazed or not, then the

site is monitored because this probably reflects a low level of grazing). • Areas currently planted in monocultural, agricultural crops (such as corn, wheat, soybeans). • Agricultural fields that are fallow and still retain evidence of fresh stubble (less than one

season since abandonment). • Grasslands, hayfields, etc. that are mowed frequently (i.e. more than three times per year). • Manicured grasslands, such as golf courses, mowed cemeteries, city parks, turf farms, or

most airfields. Each site monitored for plants is always monitored for birds. If the plant site is less than 10 acres in size then a second site that is 10 acres in size or greater is monitored for birds. The above criteria are used for selecting the second bird site, except that manicured airfields or heavily grazed pastures are acceptable for bird monitoring. This is because these habitats may harbor significant grassland bird communities. Finally, monoculture hayfields such as alfalfa and clover are also used for bird sites as these also contain substantial bird populations 4) Site evaluation, site selection, and documentation As mentioned above, with the exception of bird monitoring in wetlands and grasslands, only one site of a particular habitat type is sampled per township (although the same township may be randomly selected to monitor more than one habitat). However, the GIS identifies and ranks multiple potential sampling sites in each habitat. These sites must be evaluated, sequentially, until a site is identified that meets the specified habitat criteria and landowner access is granted to work at the site. The field crew attempts to determine ownership and gain permission to conduct monitoring at the randomly selected sampling location numbered “1” in each township. If access is granted

CTAP Protocols 6

and the habitat at that site meets the criteria mentioned above, and the site is safe and logistically practical to work in, then this location is accepted as the permanent sampling site in the township. If site number “1” is rejected for any reason (e.g. unacceptable habitat, inability to obtain landowner permission, etc.), the reasons are documented and then site “2” is evaluated, and so on until an acceptable site is found. The rest of this section describes the specific procedures followed to gain access to sites, conduct site evaluations, and document the results of these evaluations. a) Landowner access Regardless of the habitat quality of a site, monitoring is not conducted at a site without the landowner’s permission. This issue is particularly important in this project because over 90% of the land in Illinois is privately owned. Therefore, for each potential study site that is visited, ownership is determined. The landowners are contacted to gain permission to access the site and to learn about the land use history at the site. The field crew uses township land cover maps (Figs. 2, 3, 4) produced by the Illinois Natural History Survey (INHS) in conjunction with Illinois plat maps, USGS topographic maps, and the DeLorme Illinois Atlas and Gazetteer (DeLorme, 1996) to determine location and ownership of the randomly selected sampling points in each township. Although only one site will be monitored, multiple sites will often need to be visited before a site that is suitable and accessible for monitoring is located. It ultimately saves time to get ownership information for more than one site in a township before conducting site evaluations. Once land ownership is determined, the owner’s mailing address and/or phone number is identified. The field crew is usually able to locate landowners by searching a number of address databases on the Internet, county tax records, or by interviewing neighbors and relatives. Purported owners are contacted (either by phone, personal visit or sometimes by letter) to confirm their ownership, at which point the monitoring program is explained to them and access is requested. The field crew clearly states that this is a long-term program and that access is requested not only for one field season, but also for visits every five years. To document this agreement a standard letter of intent is given to the landowner (Fig. 5) Although no legal or written agreements from landowners are required by CTAP, some owners require a written release from liability in the form of a signed landowner agreement (Fig. 6). Often, for sites that are publicly owned or owned by large corporations access or research permits are required prior to monitoring. Sites such as these are identified early so as to allow adequate time for the permits to be obtained. Once access to a suitable site is obtained the assessment crew informally questions or presents a questionnaire to the landowner about historical land use practices at the site, current uses, and any plans for future uses (using guideline questions as in Fig. 7). If the landowner is not available or do not want to answer questions the questionnaire is left to be mailed later by the landowner (a stamped enveloped is provided).

CTAP Protocols 7

Decisions occasionally must be made as to when to abandon unsuccessful efforts to locate or contact landowners, or to obtain permission to access sites. As mentioned in the previous section, the field crew rejects sites if the habitat is unsuitable, unsafe, or it is unreasonable to work in for any reason. Generally, a site will be rejected if the landowner cannot be contacted during multiple attempts over a month, their requirements for permission are unreasonable, or access is denied. The point at which efforts to obtain access become excessive is subjective, but the field crew strives for consistency in their determination of when this point is reached. The underlying objective is to employ a set of protocols that will avoid rejecting sites for subjective reasons, such as perceived habitat quality. When a site is rejected, the reasons for rejection are documented. b) Site evaluations, ground truthing Most site evaluations are conducted in late winter or spring. It is logistically untenable to conduct site evaluations during the field season, but problems can also arise by evaluating sites too many months before the field season. It can be difficult or impossible, for example, to assess vegetation characteristics during the dormant season, and the longer the time interval between assessment and sampling, the greater the probability of land use changes occurring at the site (such as logging). When conducting site evaluations, the field crew gets as close as possible to the predetermined monitoring locations of the potential sampling sites as labeled on the GIS field maps. For forest and grassland sites, these locations are identified based on the best ability of the field biologists to locate the exact randomly identified coordinates depicted on the GIS field maps. These maps are often used in conjunction with USGS topographic maps and the Illinois Atlas and Gazetteer. For wetlands, the GIS randomly identifies patches of habitat (not specific coordinates) as potential sampling locations, therefore these sites are evaluated from the center of the wetland patches. The extent and condition of the habitat at each site is evaluated based on the criteria in section 3 of this report (i.e., Habitat criteria for study sites) and the site characteristics are recorded on field data sheets (Fig. 8). In addition to serving as the mechanism for selecting study sites, the process of sequentially evaluating these sites also provides information useful for ground-truthing and evaluating the GIS/land cover map. Especially in the case of grasslands, it enables CTAP scientists to assist in determining the percentage of the State that is in a condition that is at least minimally acceptable for our monitoring criteria. Therefore, even if the site is unsuitable for CTAP monitoring, the characteristics of the sites are documented. Safety and logistics of working at the sites are also evaluated during the site visit. Sites are rejected if they are determined to be unsafe or if it is logistically impractical to work on site (e.g., severe flooding, difficult access). In all habitats, if the predetermined sampling location determined by GIS is not suitable but adjacent sites are suitable, then the exact sampling locations may be shifted slightly (by the following procedures outlined below) rather than rejecting the site.

CTAP Protocols 8

c) Establishing a study plot In terrestrial habitats, if the habitat at the predetermined monitoring location (i.e., the center point in forests, the center of the baseline in grasslands, or the center of the predetermined wetland area for wetlands) does not meet the criteria for sampling then:

• If the edge of an adequately sized patch of suitable habitat is located within a 150m radius of the predetermined monitoring location, then the monitoring location is moved into the newly identified patch (a distance just sufficient to conduct the sampling).

• If there is suitable habitat in more than one direction from the original predetermined

monitoring location, then the location is moved into the closest suitable habitat.

• If suitable habitat occurs the same distance away from the location, in more than one direction, then the location is moved to the suitable habitat with the lowest compass bearing (e.g. if there is suitable habitat to the east [90o] and south [180o], the center point is moved to the east).

Sometimes only a small part of a suitable patch of habitat is within a 150m radius from the randomly selected monitoring location. In this case, the monitoring location is placed in that suitable habitat, but located at the closest distance that accommodates the minimum area required for the methods. This may result in a monitoring location that is more than 150m from the randomly selected location. If the edge of acceptable habitat is not found within 150m of the predetermined monitoring location, and/or it is not possible to gain access to the site for long-term monitoring, then that site is rejected and point numbered “2” in the same township is evaluated. This process is continued until an acceptable site is found. If none of the randomly prioritized sites in a township are acceptable, then that township is rejected and evaluation begins in the next available randomly selected township. Once the monitoring location is established, the study site is temporarily marked with flagging tape, later the site is permanently marked with metal tree anchors that are driven into the ground until they are flush with the surface (so they can not be tripped over). At a few sites owners have requested that tree anchors not be used, or that anchors not be located directly on center points. These exceptions are well mapped. Aluminum tags are tied to the anchor heads to identify the monitoring location and transect points at the site. The monitoring location and transects, as well as reference points, such as permanent fence posts, road intersections, bridges, etc., are recorded by a GPS (Global Positioning System) unit to assist in future relocation. A printout of the GPS points is used to generate a general map of the site. d) Documentation Records are kept detailing the characteristics of each site that is evaluated, regardless of whether the site is accepted or rejected. These records include information on landowner contacts (name, address, phone number, etc.) as well as site characteristics (vegetation type, obvious disturbances, etc.). All this information is recorded on a site assessment form (Fig. 8), and later added to a site identification and landowner database (see below).

CTAP Protocols 9

This process of site evaluation and documentation provides data useful for a variety of purposes: 1) Because the potential sites are selected based on information in the Land Cover of Illinois

database, the site evaluations provide a mechanism for ground-truthing the database to determine the accuracy of the land cover classification;

2) Because the randomly ranked sites are evaluated sequentially, site evaluations provide data not only on the proportion of sites that meet the criteria for acceptance in the CTAP monitoring program, but they can also be used in conjunction with information from the Illinois Land Cover database to determine the proportion of land in Illinois that meets these CTAP criteria.

3) The process of contacting landowners to gain access and evaluate sites provides information to assess the success rate and efficiency of program implementation (in terms of time investment). It also helps to learn more about the history and anticipated future use of the sites, and it provides addresses for sending landowners updated information about our activities and project results.

The following is documented for each site:

• On paper [ownership, address & phone, anecdotes about site use/history (landowner informal interview), site characteristics (e.g. age of current vegetation, vegetation type, disturbance)]

• Process of sequentially evaluating sites • Reasons for rejection if site is rejected (no permission, unsuitable) • Directions and maps to sites that will be monitored • Aerial photos (if available, e.g., TerraServer.com) • In the case of observance of threatened and endangered species, records are sent to

Heritage Database

CTAP Protocols 10

5) Plant Sampling Protocols a) Forest Sampling Protocols i) Establishing study plots Vegetation is sampled in three 1/20th ha plots (50 x 10m) in each forest patch (one forest patch per township). The plots are laid out along 50m transects that radiate out from the site’s center point at randomly selected compass bearings, starting at a distance of 10m from the center point (Figs. 9, 10). The transect bearing is determined by picking a random number between 1–360, with the constraint that no two transects can be closer together than 53°. This is to avoid overlap between the transects. When laying a transect, the tape measure used is initially pulled taut, but then allowed to lay upon the ground at all points along its length, following the contour of the ground. At both ends and at the beginning of each 10m interval, a flag is temporarily placed in the ground. The center point and the beginning of each transect (0m) are permanently marked with a metal tree anchor buried in the ground. Aluminum tags are tied to each tree anchor to identify the center point and transects. If a transect runs through a patch of uncharacteristic habitat it is relocated by choosing another random azimuth. Examples of uncharacteristic habitat include a habitat type different that the habitat type of the center point, garbage or other refuse, excavations, unnatural soil mounds, etc. Treefall gaps do not constitute uncharacteristic vegetation – they are included in the monitored transects (see Habitat criteria for study sites). If the transect crosses an interruption in vegetation, such as a stream or path etc., where more than one quadrat falls within the interruption, then the transect is terminated on the closest edge of the interruption and resumed, at the same point along the transect, on the distal side of the path. ii) GPS data A global positioning system unit (GPS) is used to record the exact latitude and longitude of the center point at each site where plants are monitored. The location of transects are also recorded (as line features). iii) General site characteristics At each site the field crew documents the general characteristics of the area around the center point of the plot. Most of these data are recorded on the data sheet in Fig. 11, and include the following: • A classification of the vegetation community based on the Natural Areas Inventory

categories (White and Madany, 1978). Categories used are shown in Table 1. • A CTAP classification of the vegetation community modified from the Natural Areas

Inventory categories. • Additional plant species not recorded during the quantitative survey. This information is

recorded on data sheets such as Fig. 12. • Brief notes describing the type and extent of obvious disturbances in the study area, defined

as a circle with a 60 meter radius from the center point. This information is used to

CTAP Protocols 11

supplement the information gained from the landowner regarding the disturbance history of the site (Fig. 7).

• The general “health” of the forest, with comments on visible evidence of disease, insect damage, pollution, drought, etc.

iv) Slope and aspect The general slope (i.e., % slope) and aspect (i.e., Azimuth) of the each transect is recorded. This is in reference to the general area that contains the transect(s) and quadrats, not necessarily the slope or aspect of the transect directly. In transects with considerable micro-topography throughout their length, an estimate or average of the overall or dominating aspect and slope conditions is provided. Slopes are measured in percent. Aspects are measured in degrees azimuth and are always taken facing downhill from the point where the slope was measured. v) Photographs Digital photos and/or 35mm slides are taken from the center point in the four cardinal directions (0o, 90o, 180o, and 270o). Photos are taken at eye level using the widest angle the camera lens will allow (generally 28mm). Pictures are taken with the highest F-stop and greatest depth of field light will allow. To avoid confusion about which site the photos depict, a photo is also taken of the site's data sheet after the habitat shots are taken. vi) Ground cover (including woody cover < 1m tall) The ground cover of vascular plants is estimated in ten 1/4m2 square quadrats along each transect (Fig. 10). The quadrats are set every 5m along the transect, starting at the 0 point. Quadrats are placed 1 meter off the transect on alternating sides. The first quadrat is always placed to the left of the transect, the next to the right, etc. More specifically, plots are placed so they cover 0.0-0.5m on left, 5.0 - 5.5m on right, 10.0 - 10.5m on left, etc., at a distance of 1.0 - 1.5m from the center of the transect. In each quadrat all herbaceous and woody (< 1m tall) species rooted inside the quadrat are recorded along with an estimate of cover for each species. To standardize cover estimates a modified Daubermire method is used (Bailey and Poulton, 1968; Abrams and Hulbert, 1987). Cover classes include: <1%, 1-5%, 5-25%, 25-50%, 50-75%, 75-95% and 95-100%. Percent cover estimates are also reported for various composite categories including total herbaceous cover (all vascular herbaceous species combined, but excluding moss and other non-vascular plants), total woody cover (for plants < 1m tall), all vegetation combined (woody and herbacous vascular plants), bare ground, leaf litter, and moss cover. In all cases, vegetation is only counted for individuals that are rooted in the quadrat, and vegetation will only be counted if it covers part of the quadrat while undisturbed. In other words, plants rooted in, but that are bent over so their cover is mostly outside the quadrat, will only be given a cover value based on the foliage that covers the quadrat where it lies naturally. Data on ground cover are recorded on data sheets similar to those in Fig. 13. vii) Woody vegetation < 5cm dbh, but at least 1m Woody plants and vine in the shrub layer are sampled in a 50m x 4m (Fig. 10) subplot centered along each transect. Each species is recorded along with a count of the stems, at least one meter tall and less than 5cm dbh (diameter breast height), rooted in the subplot. A stem is counted if it

CTAP Protocols 12

rises from ground level. Stems forking above ground level are counted as one rather than two. The stem counts for each 10m interval are kept separate as well the stem counts for the 0-1m and 1-2m on either side of the transect. These data are recorded on data sheets like Fig. 14. viii) Woody vegetation > 5cm dbh Woody plants in the tree canopy and subcanopy layer are sampled in a 50m x 10m (Fig. 10) subplot centered along each transect. Each species is tallied by recording the dbh of each stem greater than or equal to 5cm dbh. A stem is counted along the edge of the plot if at least half the diameter of the stem is within the plot. Dbh classes are used when recording the data: 5-9.9cm, 10-14.9cm, 15-19.9cm, 20-24.9cm, 25-29.9cm, 30-39.9cm, 40-49.9cm, 50-59.9cm and 60cm and above. For those >60cm dbh observers record the exact dbh measurements. Similar to the shrub subplot stem counts for each 10m interval are kept separate. Fig. 15 is an example of a data sheet used to record these data. ix) Big Plot A species list is generated by carefully searching the entire 10m x 50m area centered around the third transect (i.e., 5m on each side of the 50m-long transect, usually the third transect) and recording every species encountered. Because of time constrains searching, collecting, and identifying specimens is limited to 30 minutes in the big plot. If conditions are unsuitable (i.e. storming, darkness, etc.) this information is not gathered. Fig. 12 is an example of a data sheet used to record these data. x) Collection of voucher specimens Specimens of all plants of questionable identity are collected (when possible outside the quadrat). Each specimen is given a unique collection number on the data sheet (Fig. 12). Collection numbers are assigned by using the site identification number as the first part of the number and then sequentially numbering each specimen collected for that day as the last part of the number. For example, the first plant collected at a forest site with a township number of 506 and a site number of 2 is given a collection number of 050602F-1. Once specimens have been identified, each specimen is mounted on a herbarium sheet, labeled with the standard collection and location information (Fig.16), and deposited in the Illinois Natural History Survey Herbarium (ILLS).

CTAP Protocols 13

b) Wetland Sampling Protocols i) Establishing study plots A baseline is placed along the edge of the wetland vegetation and parallel to the long dimension of the wetland. Either long edge is used for the baseline, but most often the edge used is the side that is most accessible. The baseline length is 50m long, unless the habitat patch is less than 50m long (Figs. 9 and 17). In the latter case, the length of the baseline is the length of the habitat patch. The center of the baseline is placed at the center of the length of the wetland. A 41m transect(s) is placed perpendicular to the baseline, running into the wetland. The 0 point of the baseline is permanently marked with a metal tree anchor buried in the ground. A point is randomly selected along the baseline from which a 41m transect is placed perpendicular to the baseline, running into the wetland. When lying the transect, the tape measure is pulled taut, but laid upon the ground at all points along its length. Herbaceous vegetation is sampled in 1/4m2 quadrats at an interval of every 2m along the transect, starting 2m from the baseline. A total of 20 quadrats are sampled per site. Quadrats are placed 1m from the transect on alternate sides, starting on the left at the 2m point (e.g. the first quadrat covers the area from 2-2.5m along the transect, at a distance covering 1-1.5m left of the transect). If there is not a sufficient amount of palustrine/emergent habitat on the first transect to run the entire length (i.e. < 41m), then the field crew returns to the baseline and runs another transect from a second randomly selected point along the baseline and continues as before. Transects are terminated when they reach open water with less than 30% plant cover or when the opposite end of the wetland is encountered. Transects are placed at least 8m apart and no closer than 4m from the edge of suitable habitat to accommodate insect sampling [i.e., a maximum of 6 transects on a 50m baseline]. If the length of suitable habitat on an additional transect(s) is greater than the length needed to finish setting all the quadrats, then the field crew picks a random distance along the transect to begin setting quadrats. For example, if 12 plots are set on transect #1 and the second transect is 30m long, then the first of the remaining 8 plots is placed at a randomly selected distance of 2 -12m from the start of the transect. Plots falling into patches of uncharacteristic habitat due to a degradation factor are relocated (e.g. garbage dumped locally, excavations, unnatural soil mounds) by choosing another random number along the baseline. If the transect crosses an interruption in vegetation, such as a stream or path etc., where more than one quadrat falls within the interruption, then the transect is terminated on the closest edge of the interruption and resumed at the same point along the transect on the distal side of the path. ii) GPS data The coordinates of the baseline 0m point as well as the baseline and transects are recorded with a GPS unit. If the area of suitable habitat is about 10 acres or less then the boundary of the wetland is documented with a GPS unit. The boundary of any open water in the wetland is also recorded using a GPS unit if the extent of the open water is less than about 10 acres in size. For wetlands larger than 10 acres, part or all of the boundary coordinates may be documented with a GPS, depending on logistic constraints. In cases where GPS boundary measurements are not

CTAP Protocols 14

taken, notes are made on those parts of the wetland extending further in any direction and a site sketch is made. iii) General site characteristics Characteristics for the site encompassing the baseline and the longest transect are recorded at each wetland site and follow similar procedures to those already described for forest sites (see forest general site characteristics). The data sheet used for recording the general conditions of the wetland site is shown in Fig. 18. iv) Slope and aspect The general slope (i.e., % slope) and aspect (i.e., Azimuth) of the area containing the transect(s) are recorded. Generally this will correspond to the “tree subplot” and therefore the whole study site. Slope and aspect are measure as in Forests (see forest protocols). v) Photographs Digital photos and/or 35mm slides are taken of each site in each of the four cardinal directions (0o, 90o, 180o, and 270o) while standing on the 0m point of the baseline. For more detail, see the section on forest general site characteristics. vi) Ground cover (including woody cover < 1m tall) The ground cover of vascular plants is estimated in twenty 1/4m2 square quadrats along the transect (Fig. 17). In each quadrat all herbaceous and woody (< 1m tall) species rooted inside the quadrat are recorded along with an estimate of cover for each species. To standardize cover estimates a modified Daubermire method is used (Bailey and Poulton, 1968; Abrams and Hulbert, 1987). The following cover classes are used: <1%, 1-5%, 5-25%, 25-50%, 50-75%, 75-95% and 95-100%. In addition to estimating cover for each species individually, estimates for total percent cover for the composite categories of all species combined, all woody species, and all herbaceous vegetation are given. These data are recorded on data sheets like Fig. 19. vii) Woody vegetation < 5cm dbh, but at least 1m tall These methods are similar to those used at forest sites and the data sheet used is also similar (Fig. 20). For each species, the number of individual stems at ground level that are rooted within 2m on both sides of the transect(s) established for the quadrats are counted. The total length and portion of the transect(s) that is sampled is the same as that sampled with quadrats. Thus the total length of transect(s) sampled is 41m (i.e. a 4m x 41m area). Separate tallies are kept for the 0-1m and 1-2m distances from the transect(s). Woody vegetation data are recorded on data sheets like that shown in Fig. 20. viii) Woody vegetation > 5cm dbh A plot is established with the dimensions of the length of the baseline and the length of the longest transect (usually, 50m x 41m). Each species is tallied by recording the dbh of each stem greater than or equal to 5cm dbh. A stem is counted along the edge of the plot if at least half the diameter of the stem is within the plot. Dbh classes are used when recording the data: 5-9.9cm, 10-14.9cm, 15-19.9cm, 20-24.9cm, 25-29.9cm, 30-39.9cm, 40-49.9cm, 50-59.9cm and 60cm and above. For trees >60cm dbh, the actual dbh is recorded. Tree data are recorded on data sheets like Fig. 20.

CTAP Protocols 15

ix) Big Plot A species list is generated by carefully searching the entire 50m x 41m area (or baseline x longest transect). All encountered species are recorded. In the case that more than one transect is used then the searching area will be the length of the baseline and the length of the longest transect. Because of time constrains searching, collecting, and identifying specimens is limited to 30 minutes in the big plot. If conditions are unsuitable (i.e. storming, darkness, etc.) this information is not gathered. Fig. 21 is an example of a data sheet used to record these data. x) Collection of voucher specimens Specimens of plants of questionable identity are collected and vouchered as described in the section on forest plant monitoring methods.

CTAP Protocols 16

c) Grassland Sampling Protocols i) Establishing study plots If there is a sufficient amount of habitat suitable for sampling, the field crew sets a 50m baseline parallel to the shortest dimension of the suitable habitat patch (Figs. 9 and 17). If the patch is too narrow to run a baseline a distance of 50m, then it is still laid out as far as possible. The baseline is centered as closely as possible, on the randomly selected point (i.e. 25m on either side of the randomly selected point). For example, if the grassland is 70m wide and the random point is 15m from the edge, then the baseline would run 15m to the one edge and continue 35m on the other side rather than placing all 50m to one side. The 0 point of the baseline is permanently marked with a metal tree anchor buried in the ground. If the baseline is in the middle of a large patch of suitable habitat, then to avoid bias, a coin is flipped to determine which side of the baseline the transect should run. A 41m transect is run perpendicular to the baseline at a randomly selected distance along the baseline. When laying the transect, the tape measure is pulled taut, and then placed upon the ground at all points along its length. Herbaceous vegetation is sampled in 1/4m2 quadrats at 2 m intervals along the transect, starting at the 2m point on the transect. A total of 20 quadrats are sampled. Quadrats are placed 1m from the transect on alternate sides, starting on the left (e.g. the first quadrat covers the area from 2-2.5m along the transect, at a distance covering 1-1.5m left of the transect). If there is not a sufficient amount of habitat on the first transect to run the entire length (i.e. < 41m), then the field crew returns to the baseline and runs another transect from a second randomly selected point along the baseline and continues as before. Transects are at least 8m apart and no closer than 4m from edge of suitable habitat [i.e., a maximum of 6 transects on a 50m baseline]. If the length of suitable habitat on an additional transect is greater than the length needed to set all the remaining quadrats, then the crew picks a random distance along the transect to begin setting quadrats. For example, if 12 quadrats are set on transect #1, and the second transect is 30m long, then the first of the remaining 8 quadrats is placed at a randomly selected distance of 2 -12m from the start of the transect. In small patches where the maximum number of randomly selected transects (6) running perpendicular to the baseline would be insufficient to sample 20 plots, then the baseline is placed along the edge of the habitat patch (rather than centered on the randomly selected sample point). ii) GPS data The coordinates of the baseline 0m point as well as the baseline and transects are recorded with a GPS unit. If the area of suitable habitat is about 10 acres or less then the boundary of the grassland is documented with a GPS unit. If the area is greater than 10 acres, only part of the boundary may be assessed and, additional notes and site sketches are made to describe the further extensions of grassland. iii) General site characteristics Characteristics for the site encompassing the baseline and the longest transect are recorded at each grassland site and follow similar procedures to those already described for forest sites (see

CTAP Protocols 17

forest general site characteristics). The data sheet used for recording the general conditions of the grassland sites is shown in Fig. 22. iv) Slope and aspect The general slope (i.e., % slope) and aspect (i.e., Azimuth) of the area containing the transect(s) are recorded. Generally this will correspond to the “tree subplot” and therefore the whole study site. Slope and aspect are measure as in Forests (see forest protocols). v) Photographs Digital photos and/or 35mm slides are taken of each site in each of the four cardinal directions (0o, 90o, 180o, and 270o) while standing on the 0m point of the baseline. For more detail, see the section on forest general site characteristics. vi) Ground cover (including woody cover < 1m tall) The methods used to sample ground cover in grasslands are the same as those used in wetlands and the data sheet used (Fig. 23) is also similar. The ground cover of vascular plants is estimated in twenty 1/4m2 square quadrats along the transect. In each quadrat all herbaceous and woody (< 1m tall) species rooted inside the quadrat are recorded along with an estimate of cover for each species. To standardize cover estimates a modified Daubermire method is used (Bailey and Poulton, 1968; Abrams and Hulbert, 1987). The following cover classes are used: <1%, 1-5%, 5-25%, 25-50%, 50-75%, 75-95% and 95-100%. In addition to estimating cover for each species individually, estimates for total percent cover for the composite categories of all species combined, all woody species, all herbaceous species, all graminoid plants and all forbs are given. vii) Woody vegetation < 5cm dbh, but at least 1m tall These methods are the same as those used at wetland sites, and the data sheet used (Fig. 24) is also similar. For each species, the number of individual stems at ground level that are rooted within 2m on both sides of the transect(s) established for the quadrats are counted. The total length and portion of the transect(s) that is sampled is the same as that sampled with quadrats. Thus the total length of transect(s) sampled is 41m (i.e. a 4m x 41m area). Separate tallies are kept for the 0-1m and 1-2m distances from the transect(s). Woody vegetation data are recorded on data sheets like that shown in Fig. 24. viii) Woody vegetation > 5cm dbh These methods are the same as those used at wetland sites and the data sheet used (Fig. 24) is also similar. A plot is established with the dimensions of the length of the baseline and the length of the longest transect (usually, 50m x 41m). Each species is tallied by recording the dbh of each stem greater than or equal to 5cm dbh. A stem is counted along the edge of the plot if at least half the diameter of the stem is within the plot. Dbh classes are used when recording the data: 5-9.9cm, 10-14.9cm, 15-19.9cm, 20-24.9cm, 25-29.9cm, 30-39.9cm, 40-49.9cm, 50-59.9cm and 60cm and above. For trees >60cm dbh, the actual dbh is recorded. Tree data are recorded on data sheets like Fig. 24. ix) Big Plot A species list is generated by carefully searching the entire 50m x 41m area (or baseline x longest transect). All encountered species are recorded. In the case that more than one transect

CTAP Protocols 18

is used then the searching area will be the length of the baseline and the length of the longest transect. Because of time constrains searching, collecting, and identifying specimens is limited to 30 minutes in the big plot. If conditions are unsuitable (i.e. storming, darkness, etc.) this information is not gathered. Fig. 25 is an example of a data sheet used to record these data. x) Collection of voucher specimens Specimens of plants of questionable identity are collected and vouchered as described in the section on forest plant monitoring methods.

CTAP Protocols 19

6) Terrestrial Insect Sampling Protocols Understory arthropods (with a primary focus on insects) are sampled in each terrestrial habitat at the same times and locations where plant sampling takes place. Information on establishing study plots and collecting location information and site characteristics are found in the previous sections on plant sampling (see Figs. 9, 17, and 26). a) Sampling locations Sampling involves the collection of insects on two parallel transects at each site. In the forest habitat, insect transects are placed 3m to each side of one of the three, 50m-long plant transects. Usually the first transect (the three plant transects are arbitrarily numbered) is chosen, but sometimes the insects are collected on the transect with vegetation most suitable for insect sampling. In wetland and grassland habitats there is only one, 41m-long plant transect, but in some cases the study site is so small that the plant transect must be broken into more than one segment. In these situations, the insect sampling is still generally conducted 3m to each side of the transect, with data collection interrupted between the segments. If this process is to laborious depending on the vegetation structure an alternate sampling area may be identified (e.g., a line parallel to the baseline on side opposite to transects). In these situations, path of travel and area colleting are recorded (see next section). b) Sampling methods The collection methods used in each habitat are identical. Two standard sweep net samples are collected at each site and data about the sampling are recorded on data sheets similar to those shown in Fig. 27 (forest sites) and Fig. 28 (wetland and grassland sites). The collections are standardized by making 100 sweeps of the net in each transect sample. A stroke is one swing of the arm, to the left or right, in front of the collector as he or she walks forward. Usually, each swing of the arm would be accompanied by a step, so an insect transect is about the same length as the plant transect – 40 to 50 meters. However, it is the number of sweeps that is standardized rather than the distance sampled. The height and length of the sweep varies with the vegetation encountered, but usually the lowest part of the arch would sweep approximately along the top 1/2 meter of vegetation and the highest part of the arch would be about shoulder height. Once the net starts swinging it is kept in continuous motion so insects will not escape. In situations where the plant transect is broken into more than one segment, the net is swept to the end of the first transect segment as described above, at which point the net is twisted closed to hold the arthropods already captured. The sampler then travels to the next transect segment and sampling continues until 100 sweeps are made. The net is easily swept through most vegetation, but when vegetation is too prickly (i.e. Rosa multiflora, Rubus sp., or Ribes sp.), too dense, or too tall to sweep through, then the sweep path deviates around these into vegetation more suitable for sweeping. This may mean that the sweep path deviates away from the transect for a few meters, or that an alternate transect (or transect segment), other than the first is swept. On occasion, another location (which may or may not be a linear path) within the study other than a transect, such as unmowed strips or along the baseline in wetlands and grasslands, is swept. The study site is defined as the area forming a circle of

CTAP Protocols 20

75m radius in forests and the rectangular area formed by the length of the baseline and longest transect in wetlands and grasslands. Once the net is swept through the vegetation 100 times, the open end is quickly twisted closed with the hands. The insects inside the net are shaken into PTUIEs, Photo Tactic Utility Insect Extractors. The PTUIEs are used to separate insects from duff (leaves, seeds, sticks, etc.) in the field while the insects are still alive. A PTUIE consists of a large plastic jar kept dark inside by applying layers of paint or black plastic and duct tape on the outside of the jar. A hole is cut into the lid of the jar. A straight sided, clear, 16oz.-soda bottle, with a corresponding hole in its side, is riveted and sealed with a hot glue gun to the lid of the large plastic jar. A whirl pac (plastic bag), one third full of 70% ethyl alcohol, is attached to and hangs down from the mouth of the soda bottle (Fig. 29). Insects are attracted to the light passing through the clear soda bottle, then drop into the ethyl alcohol in the whirl pac attached to the mouth of the bottle, thus separating themselves from the duff. Insect sampling begins as soon as the plant transects of a site have been laid out. This allows the PTUIEs to be set up as long as possible while plant monitoring occurs. To begin the process, two whirl pacs are labeled with collection numbers, filled about one third full with ethyl alcohol and attached to the two soda bottles, which in turn are attached to the jar lids of the two large plastic jars. The two soda bottle assemblies and the large plastic jars are place separately in a shady place, safe from wind and foot traffic. The insect sample from the sweep net is placed inside the large plastic jar of the PTUIE. The soda bottle assembly with the jar lid is replaced quickly without dumping the alcohol from the whirl pac into the jar. The assembled PTUIE is placed in the shade and on a stable surface that will allow the whirl pac to hang down lower than the large jar. It is allowed to sit a minimum of 30 minutes to allow the insects to move from the dark of the large plastic jar into the light of the soda bottle, and drop into the alcohol. Optimally, the PTUIE is allowed the sit until there is no more insect activity. This time length is effected by how many insects are in the sample and the type of duff they are moving through. The amount of time the PTUIEs are allowed to sit is recorded on the field sheet. Once the sample is collected, the whirl pac bag is removed. The sample number is written on paper, with pencil, and placed inside the whirl pac. The top edge of the whirl pac is folded over three times and twist tied closed to create a nearly leak proof seal. In their constant pursuit of food, spiders sometimes set up web traps in the lid opening, preventing them and other arthropods from dropping into the alcohol. The PTUIEs are picked up and gently tapped to nudge them down the soda bottle neck. Once the insects have stopped coming out of the dark jar and have dropped into the alcohol the PTUIE can be opened to remove the duff. An examination of the jar will reveal arthropods that never left the darkness. They will include small homopterans, which stick to the sides of the jar because of condensation. There will also be moths and other nocturnal arthropods. This failure to extract every arthropod swept into the net is acceptable in quantitative sampling because it is a consistent variable from sample to sample. The insects that remain in the jar are emptied with the duff. The PTUIEs are

CTAP Protocols 21

thoroughly cleaned out at the next site when the condensation has dried and any remaining material that was stuck to the sides of the jar is removed. After processing, all the samples are stored in vials of 70% ethanol and maintained for future study and possible inclusion into the permanent INHS insect collection. Finally, if the vegetation is too wet due to rain nearly before plant sampling, insect sampling is not conducted. c) Notes on sampling methods Sweep sampling collects arthropods that travel on leafy understory vegetation. It under-samples terrestrial and bark dwelling insects, as well as many fast aerial fliers and canopy dwelling insects. Many arthropods have pronounced circadian activity patterns, so this method also under-samples most nocturnally active species. However, sweep sampling provides a good, easily quantifiable collection of arthropods and can be easily employed in all the habitats CTAP biologists are monitoring. d) Collection numbers Each collection is given a unique collection number on the data sheet (Fig. 27 and 28). Collection numbers are assigned by using the site identification number as the first part of the number and then attaching a number followed by the letter "I" to the site identification number. The collection from the first sweep, usually on the left side of the transect, is given the number one and the second, usually the right side of the transect, is given the number two. For example, the first sweep, at a forest with a site identification number of 050602F is given a collection number of 050602F-1I (see plant methods for an explanation of the generation of site identification numbers).

CTAP Protocols 22

7) Bird Sampling Protocols a) Requirements and yearly preparation To be qualified to conduct CTAP bird monitoring, members of the CTAP field crew must be familiar with the songs and calls of all bird species likely to be encountered in Illinois. Each spring the observers review tape recordings of the calls and songs of Illinois species. Because about 90% or more of the birds detected are not seen, it is also critical that observers have good hearing. Hearing loss will seriously affect the census results. Unlike sampling of other taxa, samples cannot be collected for later verification. Not only is it important to be able to identify all birds detected, but it is also important to be able to accurately estimate the distance to each bird detected. Small differences in estimated distances can lead to very different estimates of population density, thereby leading to higher variance estimates and lower statistical power to detect trends. This means that inter-observer differences in distance estimates could mask real trends in population density. Therefore, yearly efforts must be made to control for observer differences by having the observers spend time together calibrating their distance estimates. This is particularly important because the project is expected to continue longer than the tenure of any given member of the field crew. Quality control is a critical issue for CTAP bird monitoring, so the importance of yearly calibration cannot be overestimated. Calibration of distance estimates is accomplished by having the field crew visit habitats of various densities. At each site the crew independently estimates the distance to various trees, shrubs, or other landmarks. They then compare their estimates to the actual distances as determined by a calibrated laser rangefinder. They continue this process until each of their estimates for distances of less than 50m are accurate to within 5m and their estimates between 50-100m are accurate to within 10m. Moreover, they should have at least 90% accuracy at estimating if distances are greater than or less than 50m, 75m, and 100m. This process is then repeated by estimating the distances to actual birds seen, and then birds that are heard but not seen (but whose location can be confirmed afterwards). Calibration takes some practice because different species are louder than others, and even the density of the vegetation or the humidity can effect how well sounds travel (in other words, how quickly signals decay), which effects perceived estimates of distance. Finally, in addition to estimate distances, the observer has to be able to determine the direction of each bird species. Knowledge in the use of a compass is also essential. b) Establishing census points The size of the habitat patch where the study sites are located will vary. Therefore the number of bird census points that can be fit into a site may vary. Because of this, the number of points censused at a site is not standardized. A minimum requirement, however, is that one census point is always located at the center point of the site (the randomly selected point where plant monitoring is centered). Although the center point (CP) is the only required census point and overlaps with the botanical center point, additional census points (up to a maximum of about 15 points) may be added. Once

CTAP Protocols 23

added, all these points should be censused during each subsequent census. These additional census points should be numbered consecutively (1,2,3, etc.). In large forest patches these census points can be established along existing trails at 150m intervals. In small patches the points may be scattered through the forest (at 150m intervals) in whatever locations will maximize the number of points that can be fit into the forest patch. However, no point should be closer than 150m from any other point, and a point is made to stay 50 or more meters from the forest edge. When establishing these points, if time allows, the field observer locates the approximate census locations on the field map before entering the forest, and then attempts to find those predetermined sites by visiting the site. This will help avoid biasing the census by establishing points at what appear to be the best sites for birds. When establishing forest census points, it is important to stay within the forested habitat, however, small areas of other habitat types within the forest (e.g. small wetlands, areas of second growth) are still acceptable. At wetland and grassland sites, similar procedures are used for placing census points, however, the points are not required to be 50 m from the habitat edge. This would be impossible in many cases because these habitat patches tend to be small in Illinois (see Grassland habitat criteria). In wetlands and grasslands sites, census points are placed 300 m apart due to the more open habitat. A scouting trip is generally made to each site the day before conducting the census to find the center point and establish any additional census points. This saves precious census time in the morning. Once census points are established, these same locations will be used for all subsequent censuses. Therefore, the location of the points should be accurately labeled on the field maps so that they can be located during subsequent visits. Since census points will not be permanently marked in the field, it is important to make notes of any information that will help to relocate them. For example, observers record the distance and direction to each point from the previous point, and they occasionally record the distance and direction to large trees, bends in a stream, trail intersections, or other recognizable landmarks. Field notes should be detailed enough so that another person would be able to find these sites. In addition, to the all field notes and maps for subsequent census points relocation, all point are GPSed. c) Recording habitat data Detailed vegetation data are collected at the center point by CTAP botanists. However, we would also like to know something about the habitat at each additional bird census point. CTAP ornithologists take note of the habitat type at each census point using the same classification as the botanists (Table 1). d) When to census Censuses should be conducted during the period of peak breeding activity when territorial males are singing, but after the spring migration period is mostly completed. In Illinois the acceptable period for censusing generally lasts from the last week of May through the third week of July, however, yearly weather fluctuations may shift this period slightly earlier or later. Because of the large latitudinal range encompassed by Illinois, it is most efficient for sampling to begin

CTAP Protocols 24

earlier and end earlier down south, so an attempt is made to start in southern Illinois and proceed north until the breeding season has “caught up” in more northern areas of the state. On any given day, the first point of a census (usually the center point) – should be censused as close to sunrise as possible (once it is light enough to identify birds visually). Because bird activity can drop off dramatically as the day progresses, the last point count should be completed no later than 4-41/2 hours after sunrise. No census point should be initiated later than 10:00 am, or should continue past 10:30 am, with the possible exception of 11:00-11:30 am on cool, overcast days. e) Acceptable weather For data to be comparable among census points and among years, censuses must be conducted under favorable weather conditions. There should be good visibility with little or no precipitation and light winds. Occasional light drizzle may not affect bird activity, but censuses are not conducted when there is heavy fog, steady drizzle or rain, or when wind speed exceed 12 mph (Beaufort scale of 4-5 – see Table 2. Weather conditions, including noise levels, are recorded for each census point. f) Conducting point counts Each point count lasts 10 minutes. From a stationary location at the census point, the observer records all (but only) the birds detected by sight or sound during the 10 minute period. The data are recorded directly onto the data sheets - tape recorders are not used for later transcription. A watch (preferably a stopwatch) is used to keep track of the time. The data sheet (Fig. 30) also provides space to record when the 3 minute, 5 minute, 6 minute, and 8 minute marks are reached during the count (see example data sheet). This is because monitoring programs often differ in the length of their censuses, and this will facilitate comparisons to those studies. When conducting the point counts, absolutely no coaxing is allowed. Observers do not “spish”, imitate calls, or use playback recordings: they remain entirely silent during the counts. Observers record the direction and estimated distance to each bird, being particularly careful to note which birds are detected within or beyond 50 meters of the census point. Usually the initial distance to a bird is recorded. If however, the bird was initially farther away than 50 m, but subsequently moved closer than 50 m, the closer distance is recorded. Often multiple individuals of one species will be detected. To help keep track of the number of individuals detected during a count, a column is provided on the data sheet to record the direction to each bird. The direction information is also helpful for determining if a bird heard at one census point is the same as one heard at a previous point. (For example, a bird heard 80 m to the north at one point may be the same as one heard 70 m to the south at the next point, if the next point is 150 m north of the first point). If the observer is absolutely certain a bird detected at a point count is the same individual detected at a previous point, then the bird is not counted again. Normally, the bird is recorded on the point that it is closer to. The data sheet also provides space to record if the birds are or are not actually in the focal habitat patch. For example, observers may detect Turkey Vultures or Chimney Swifts flying over the

CTAP Protocols 25

forest, and loud birds such as Ring-necked Pheasants or American Crows in surrounding agricultural fields may be detected from within forest patches. Because these birds are not in the focal habitat patch, they are coded differently (Table 2). Birds flying over the census point should still be reported as occurring within 50 meters if they are within 50 m of the census point at any time while they are flying over. For birds occurring in large flocks, a column is provided to record the number of individuals detected at a given distance and direction (Fig. 30). Finally, although the purpose of the point counts is not to provide an inventory of birds at a given site, interesting birds are often detected before or after a point count is conducted, or while walking to the points. Even though these birds are not recorded on the point counts, they often provide important information about a site. Therefore, the data sheet provides space to record these additional bird sightings. g) Playback tapes in wetlands Point count censuses do not provide an equal opportunity for detecting all species of birds. Certain groups of birds will be under-sampled, including nocturnal species such as owls and nightjars, gallinaceous species such as turkeys, species that are often quiet (such as raptors), and other species that are often cryptic, such as some wader and other marsh birds. Using playback tapes in wetlands helps increase the probability of detecting many wetland species. Therefore, in wetland habitats, an additional census period is conducted after each point count is completed. Fourteen wetland species (Table 3) expected to be found in many “healthy” Illinois wetlands have been recorded on a cassette tape for a duration of approximately 1 minute followed by a pause of approximately 1 minute. This tape is played at each wetland census point and the response to any species on the tape is documented. h) Additional grasslands and wetlands for bird monitoring The site monitored for plants is always monitored for birds. For grasslands, if the plant site is less than 10 acres in size then a second site, if available, that is 10 acres in size or greater is monitored for birds. The same criteria used in the Grassland habitat criteria section are met for the second bird site, except that manicured airfields, some monocultures such as alfalfa and clover, or heavily grazed pastures are acceptable for bird monitoring. This is because these habitats may harbor significant grassland bird communities even though the sites are highly manicured and therefore not suitable for plant monitoring. In these grasslands point counts are conducted as previously described. In the case of wetland sites, many of the primary sites are small and degraded. If other, generally the largest, wetland sites are available in the target township, then some of those sites may also be monitored for birds. This depends to some extent on the time available. Extra bird sites are monitored using the same protocol as for the primary wetland sites, including the use of the playback tape.

CTAP Protocols 26

II - Aquatic monitoring protocols: Ephemeroptera (mayflies), Plecoptera (stoneflies), and Trichoptera (caddisflies) (collectively, EPT taxa). 1) Selection of Random Stream Locations Stream reaches were chosen for sampling at random from two scales. The first scale is at the level of the township, of which over 1700 exist in Illinois. CTAP chose 150 townships from this population (Fig. 31). The Illinois Streams Information System (ISIS, 1999; Fig. 32), a digital database of Illinois streams, was used as the basis for identification of stream reaches within a township. The database includes streams draining areas greater than 10 mi2, based on USGS 1:100,000 scale data. Each stream is represented by discrete segments beginning and ending at the Public Land Survey section lines (those that demark sections on topographic maps) or at the confluence of streams. Stream segments contained within each township were randomly ranked to establish sampling priority (Fig. 33). Those with the lowest rank number were visited first to access them for suitability of sampling. 2) Acceptability of Stream Segments Stream segments are acceptable for use if the following criteria are met: • Reasonable access by road or short footpath is available. Moderate amounts of sampling

gear are necessary for this work. Therefore, relatively short distances to the stream are advantageous for finishing sampling protocols in a timely fashion. Sites should require ≤30 min. walk to reach a sampling location from an existing road or footpath.

• The stream segment must afford safe entry into the streambed. Safe parking must be available. Moreover, streams must be wadeable to be safely and effectively sampled.

• No stream segment should be sampled immediately below a wastewater treatment outfall, or where a strong smell of sewage is evident.

• Moderate-to-severe flooding of a stream segment removes a segment from immediate consideration. This segment may be scheduled for a latter visit when floodwaters have receded. Alternatively, other suitable segments may be sought within the township.

• Sampling of the stream segment must be accomplished within a suitable biological window for the assemblage in question.

• The stream segment should be sufficient in size to support aquatic life throughout most years. Occasionally, no stream segments in a selected township hold water throughout the year. These are then rejected entirely, and another township is chosen and assessed for suitability.

3) Establishing a Sample Reach Once a stream segment meets the above criteria, a specific sample reach is established. Frequently, more than one suitable access site is available due to multiple access points. Some guidelines are provide below that minimized bias in selecting the reach:

• The downstream-most road crossing is chosen whenever possible. This provided samples that reflected the whole of the segment, and increased the likelihood of flowing water throughout the year.

• Reaches on small streams are established at sufficient distances from the confluence with larger ones to avoid any influence of the latter (i.e. due to flooding).

• Reaches are established in habitat that is prevalent for the segment.

CTAP Protocols 27

• Reaches are established no closer than 20 m from bridges, large culverts, or major drainage tiles. These structures may influence the local community by producing scouring of the streambed.

• Only naturally occurring substrates are sampled. Sampling of unnatural substrates (i.e. road building materials) where no coarse mineral substrates naturally occur may bias samples.

Sample reaches are approximately 100 m in bank length (meanders included). This distance is generally sufficient to provide a diversity of habitats including riffles, undercut banks, pools, and wood debris snags. 4) Geographic Referencing of the Sample Reach Location information is recorded for the reach from 1:24,000 scale maps and from DeLorme Illinois Atlas and Gazetteer (DeLorme, 1996). Geographic coordinates are recorded on site using one of several makes of global positioning system or by obtaining coordinates from Delorme’s® CD Street Atlas USA 2.0 (DeLorme, 2000). These locations are generally recorded from the middle of the reach. The following location information is gathered for each reach: • seven digit unique site code • county • stream name (if unnamed, then recorded as

Unn. trib. of the nearest parent stream) • nearest permanent, small municipality (e.g.,

not Chicago or Springfield), straight line distance (in km) from center of town to reach, and 8 point compass direction from town

• principal meridian, township, range, and section numbers

• latitude and longitude in decimal degrees to 4 significant digits, longitude as negative number

• Illinois township number, rank, stream segment number

5) Photographs Photographic documentation occurred at each sample reach to document landuse. A minimum of two photographs, usually facing upstream and downstream from the center of the sample reach, were taken. These photographs are archived in digital format to document changes in landuse over time. 6) Aquatic Insect Sampling Methods - Ephemeroptera (mayflies), Plecoptera (stoneflies), and Trichoptera (caddisflies) (collectively, EPT taxa) a) Phenology of Sampling Burks (1953), Frison (1935), Ross (1944), and INHS insect collection databases are consulted for information on the phenology of EPT species in Illinois. April 1 through 15 May provided the greatest diversity of EPT species. Even though adults of many of these species can be taken later in summer, the immature specimens (the object of sampling efforts) would be largely guaranteed to still inhabit streams across Illinois within this time frame. To control for differences in phenology along the great latitudinal span of the state, sampling began in southern Illinois and proceeded northward.

CTAP Protocols 28

b) Sampling Protocols EPT are sampled using a standardized, semi-quantitative, multi-habitat approach. This approach is designed to capture a large proportion of the EPT species found in wadeable streams, while also weighing time spent at each site and resources available for processing of samples. In Illinois streams, the greatest proportion of EPT species are taken in two general classes of habitat: high energy and low energy. High-energy habitats occur where water tumbles over hard substrates. These substrates included coarse gravel and cobbles (riffles) and woody debris (a.k.a. snags). In Illinois, snags and riffles supported similar EPT species, and are viewed as interchangeable. Some streams have abundant riffles, while others have snag habitat (where sand, clay, or silt were the predominant channel substrates). Low energy habitats included undercut banks, pools, vascular plants, and shallow runs. These support a subset of the riffle assemblage and fauna typical of slow water and root zone at the water/land interface. Undercut banks are the preferred low energy habitat type as they provide the most slow-water taxa. Hence, in most streams, riffles or snags and undercut banks are sampled. Occasionally, neither riffles nor undercut banks are available (as in recently ditched agricultural streams). In that case, runs provided the only habitat to sample.

i) High-Energy Habitats Two riffle samples are taken from two separate riffles within the reach. The

sample area is standardized to the dimensions of the dipnet bag (34 X 45 cm) superimposed upstream of the dipnet. Larger mineral substrates are washed to dislodged taxa into the net. Cobbles are inspected visually for tightly adhering taxa, then discarded. Finer substrates are turned by hand, then kicked using the foot to dislodge taxa deep in the sediments. Two snag samples, limbs 2-10 cm in diameter and approaching 3 m in total length, are collected from flowing water areas. Entrained leaf packs associated with snags are not partitioned from the wood. The wood and associated debris are dislodged into the net. Large debris are inspected for cryptic taxa, washed, and discarded until a relatively small volume remains in the net.

ii) Low-Energy Habitats

Two bank samples are collected from current-swept banks where the exposed roots of trees or grasses are abundant. The dipnet is thrust to the bottom of the bank into fine sediments, which are disturbed by foot. The net is progressively moved up the bank in the rooted zone and substrates variously kicked and pulled free into the net. Two aquatic vascular plant samples are collected when no bank habitat is available. This is accomplished by pulling all vegetation from a 34 X 45-cm area. These are placed in the net for latter inspection. Kicking of the sediments in the sample area dislodged organisms that flowed into the net.

c) Habitat Quality Assessment Values for over 50 variables are recorded during each stream visit. This information, recorded in a standardized form (Fig. 34), tracks information as varied as location, date and time, stream identity and size, sediment and water characteristics, and a 12-point habitat quality rating system. This form is adopted from various USEPA documents (Barbour et al. 1999 and Plafkin et al. 1989). Conditions are assessed for the 100-m stream reach only.

CTAP Protocols 29

d) Water Chemistry All water chemistry and temperature values are obtained using a Solomat 520-C multiparameter meter. This instrument is calibrated daily as per manufacturer instructions. Measurements are taken from upstream of sampled habitat to ensure that sampling activities do not interfere with instrument readings. e) Sample Processing, Vouchering of Specimens All EPT specimens are picked while in the field. Sample debris from each replicate is placed in a white tray and flooded with water. Specimens are deposited in 80% EtOH. All EPT specimens are identified to species where possible, stored in separate vials, and labeled with the location information summarized above. These vials are deposited in the INHS insect collection as a voucher of the taxa present at each site.

CTAP Protocols 30

III - Literature Cited Abrams, M.D. and L.C. Hulbert. 1987. Effect of topographic position and fire on species

composition in tallgrass prairie in northeast Kansas (USA). The American Midland Naturalist 117:442-445.

Admiraal, A. N., M. J. Morris, T. C. Brooks, J. W. Olson, M. V. Miller. 1997. Illinois Wetland

Restoration and Creation Guide. Illinois Natural History Survey, Special Publication 19. pp. 188.

Bailey, A. W. and C. E. Poulton. 1968. Plant Communities and Environmental Interrelationship

in a Portion of the Tillamook Burn, Northwestern Oregon. Ecology 49:1-13. Barbour, M. T., J. Gerritsen, B. d. Snyder, and J. B. Stribling. 1999. Rapid Bioassessment

Protocols for Use in Streams and Wadeable Rivers: Periphyton, Benthic Macroinvertebrates and Fish, Second Edition. EPA, 841-B-99-002. U.S. Environmental Protection Agency; Office of Water; Washington, D.C.

Burks, B. D. 1953. The mayflies, or Ephemeroptera, of Illinois. Bulletin of the Illinois Natural

History Survey 26: 1-216. DeLorme. 1996. DeLorme Illinois Atlas and Gazetteer. 2nd edition, Freeport, Maine. DeLorme. 2000. Topo USA 2.0: DeLorme’s® CD Street Atlas USA 2.0. DeLorme, Yarmouth,

Maine. Frison, T. H. 1935. The stoneflies, or Plecoptera, of Illinois. Bulletin of the Illinois Natural

History Survey 20: 1-471. Illinois Department of Natural Resources. 1995. Land Cover of Illinois Database. An electronic

database. http://www.inhs.uiuc.edu/igis/illinois/index.htm Illinois Department of Natural Resources. 1996. Illinois land cover, an atlas. Illinois Department

of Natural Resources, Springfield, IL, IDNR/EEA-96/05. Illinois Natural Resources Geospatial Data Clearinghouse. 1997. Illinois Wetlands Inventory

database. An electronic database. http://www.isgs.uiuc.edu/nsdihome/ISGSindex.html The Illinois Streams Information System (ISIS). 1999. User’s Manual: Illinois Stream

Information System. Illinois Department of Natural Resources, Office of Resource Conservation, Springfield, IL.

Plafkin, J. L., M. T. Barbour, K. D. Porter, S. K. Gross, R. M. Hughes. 1989. Rapid

Bioassessment Protocols: or Use in Streams and Rivers: Benthic Macroinvertebrates and Fish. EPA, 444/4-89-001. U.S. Environmental Protection Agency; Office of Water; Washington, D.C.

CTAP Protocols 31

Ross, H. H. 1944. The caddis flies, or Trichoptera, of Illinois. Bulletin of the Illinois Natural History Survey 23: 1-326.

Suloway, L. and M. Hubbell. 1994. Wetland Resources of Illinois: An analysis and Atlas.

Illinois Natural History Survey, Special Publication 15. pp. 88. White, J. and M. H. Madany. 1978. Classification of natural communities in Illinois. Pages 310-

405 (Appendix 30) in J. White, Illinois natural areas inventory technical report. Volume 1: Survey methods and results. Illinois Natural Areas Inventory, Urbana, IL.

CTAP Protocols 32

IV - Tables and Figures a) Terrestrial monitoring protocols i) Figures Fig. 1. Location of the first 50 randomly selected townships

a) forest b) wetland c) grassland

Fig. 2. Land cover township map – forest Fig. 3 Land cover township map – wetland Fig. 4. Land cover township map – grassland Fig. 5. Landowner agreement form Fig. 6. Liability form Fig. 7. Land use survey Fig. 8. Data sheet – site evaluation Fig. 9. Transect setup – forest, wetland, and grassland Fig. 10. Forest plot layout along transect Fig. 11. Forest plant data sheet – general conditions Fig. 12. Forest plant data sheet – site species list/big plot Fig. 13. Forest plant data sheet – ground cover Fig. 14. Forest plant data sheet – saplings/small tress:shrubs/vines Fig. 15. Forest plant data sheet – tress Fig. 16. Herbarium specimen with voucher information Fig. 17. Wetland and grassland plot layout along transect Fig. 18. Wetland plant data sheet – general conditions Fig. 19. Wetland plant data sheet – ground cover

CTAP Protocols 33

Fig. 20. Wetland plant data sheet – woody vegetation Fig. 21. Wetland plant data sheet – site species list Fig. 22. Grassland plant data sheet – general conditions Fig. 23. Grassland plant data sheet – ground cover Fig. 24. Grassland plant data sheet – woody vegetation Fig. 25. Grassland plant data sheet – site species list Fig. 26. Insect sweeping layout along transect Fig. 27. Insect data sheet – forest Fig. 28. Insect data sheet – wetland/grassland Fig. 29. Photo Tactic Utility Insect Extractors (PTUIEs) Fig. 30. Bird census data sheet ii) Tables Table 1. CTAP community categories Table 2. CTAP bird census codes Table 3. List of CTAP marsh birds b) Aquatic monitoring protocols i) Figures Fig. 31. Location of the first 50 randomly selected townships for streams. Fig. 32. Ten Illinois Streams Information System (ISIS) basins. A regionalization scheme for CTAP stream data. Fig. 33. Township 001401S (near Antioch, IL) with Land Survey Sections, towns, lakes, and streams. Note that stream segment numbers change as streams cross section line, or as they unite with other streams. Segments 1-3 are of highest priority assessment and monitoring. Fig. 34. Habitat quality assessment form. Values for over 50 variables are recorded during each stream visit on this form.

CTAP Protocols 34

V - Appendix

Logistics for using EPT to indicate stream health

CTAP assessment of wadeable streams is heavily based on the condition of biological communities. Assessment of near-stream habitat quality and measurement of several in-situ chemical and physical parameters provide additional that may help to identify specific stress agents causing impact. Barbour et al. (1999) discussed some of the strengths of using biological communities to assess stream health: • They are a reflection of the ecological integrity (the protection of which is a primary goal of

the Clean Water Act) of the stream • They integrate the effects of different stressors, providing a broad measure of their aggregate

impact • They integrate stressors over time and provide an ecological measure of fluctuating

environmental conditions • When compared to the cost of assessing toxic pollutants, sampling of biological communities

can a be cost effective • The health of some biological communities is of direct public interest (e.g., commercial and

sport angling) • Biological communities may provide the only practical means of evaluating some impacts

due to the lack of specific criteria (e.g., habitat degradation due to flooding caused by change of hydrologic regime)

Aquatic Macroinvertebrates as Indicators of Stream Condition

The use of aquatic macroinvertebrates (in most situations, dominated by insects) as indicators of water quality has increased dramatically in the past two decades, and the widely recognized effectiveness of this assemblage for detecting impairment in streams and rivers ensures its continued use (Davis and Simon 1995, Loeb and Spacie 1994, Barbour et al. 1999, Rosenberg and Resh 1993b). Several reasons for using aquatic macroinvertebrates are summarized from Rosenberg and Resh (1993a) and Barbour et al. (1999):

• They occur in all streams and within nearly all microhabitats • A large number of species offer a wide range in responses to environmental stresses • They are sedentary, permitting effective spatial analysis of pollutants and disturbance effects • They have long life cycles, allowing investigation of temporal changes cause by

perturbations. • Sampling requires few personnel, inexpensive gear, and produces only minimal, short-term

impacts upon the community • They are the food base for most vertebrates found in streams. EPT as Indicators of Stream Condition: In order to reduce the cost and effort associated with sampling the entire assemblage, CTAP has elected to use three orders of aquatic insects as indicators of condition: the Ephemeroptera (mayflies), Plecoptera (stoneflies), and Trichoptera (caddisflies) (collectively, EPT taxa). These often contribute the major proportion of the

CTAP Protocols 35

abundance and species richness to the aquatic macroinvertebrates assemblage found in streams. EPT taxa richness (number of unique types in a sample, but may be identified only to the genus level), is one of the most efficient indices of stream condition. The history and usefulness of EPT taxa was recently summarized by Lenat and Penrose (1996). Lenat (1993) found that quality ratings based on the EPT index varied predictably across Mountain, Piedmont, and Coastal Plain Ecoregions in North Carolina. Wallace et al. (1996) reported that EPT richness correlated well with several measures of stream ecosystem function (e.g., nutrient processing) and demonstrated that it could assess habitat-specific impact. Barbour et al. (1992) stated that EPT taxonomic richness varied much less than total invertebrate richness, density, or biomass estimates. They concluded that the EPT index was relatively easy to obtain, and that it was one of the simplest indices for non-biologists to use and understand. Additionally, numerical disturbance/pollution tolerance values, indicating the relative sensitivities, exist for many EPT (and other macroinvertebrates) taxa resident in the upper Midwest (Hilsenhoff 1987) and are summarized for elsewhere in the U.S. by Barbour et al. (1999). Additional reasons for adopting EPT taxa is the INHS’ long and distinguished history of research on the systematics, ecology, and distribution of these insects. State identification manuals exist for all three orders (Burks 1953 for mayflies, Frison 1935 for stoneflies, and Ross 1944 for caddisflies). These were the benchmark works of their time, and in some instances still serve as the regional standard. Most specimens associated with these statewide treatments still reside in the insect collections of the INHS. This allows confirmation of specimens by directly comparison to type or authoritatively identified specimens. Data capture of some 710,000 EPT specimens has just been completed, permitting a rapid comparison of present data with that collected before the worst degradation of Illinois streams took place. These databases have increased the efficiency of evaluating of losses in EPT species across the state (DeWalt et al. 2001, DeWalt et al. 2002, Webb and Harris 1993). A web-based EPT database is available at www.inhs.uiuc.edu/cbd/EPT/index.html. Ecological Indicators Derived EPT Assemblages: EPT samples do not automatically constitute environmentally informative data. While knowing the environmental requirements of certain species may lead to their use as indicator organism, often there is not enough information about each species to use them as a predictive tool. Most stream biologists today rely upon numerical ecological indicators, with known statistical properties, to make sense of sample data. Below is a summary of the metrics and multimetric indices used by CTAP to monitor stream condition.

EPT Species Richness decrease Taxa Dominance (proportion of individuals devoted to single most abundant taxon)

increase

Hilsenhoff Biotic Index (HBI) increase Physical/Chemical Parameters

Habitat Quality Index decrease Temperature mostly increase Dissolved Oxygen decrease/increase PH decrease/increase Conductivity increase

CTAP Protocols 36

Habitat destruction and degradation drastically reduce livable space for aquatic organisms. This, coupled with nutrient enrichment and siltation from farm fields and residential and industrial construction sites has done the most damage to aquatic systems (Karr et al. 1986). Measurements of habitat quality are important in estimating the potential for healthy aquatic communities in a stream system. For example, an increase in water temperature may indicate agriculture land or an urban area near the stream. This may lead to the loss of cool water species. Dissolved oxygen in Illinois streams can vary from supersaturated conditions brought on by heavy algal blooms during daylight hours, to hypoxic conditions in the early morning hours. The latter may result from high loadings of organic material from human and livestock sources, from demand by the heavy blooms of algae, or may occur naturally due to slow flow or autumnal leaf fall. Low oxygen may limit some species presence in a stream, especially those lacking well developed gills and summer diapause of eggs or larvae. pH generally is near neutral in most Illinois streams. Some mine drainage can depress pH in coal mining areas of the state. High pH can result from increased photosynthesis in agricultural (cleared riparian zone) streams, through dissolution of naturally occurring calcareous bedrock, and through industrial wastes with high pH. Literature Cited Barbour, M. T., J. Gerritsen, B. d. Snyder, and J. B. Stribling. 1999. Rapid Bioassessment

Protocols for Use in Streams and Wadeable Rivers: Periphyton, Benthic Macroinvertebrates and Fish, Second Edition. EPA, 841-B-99-002. U.S. Environmental Protection Agency; Office of Water; Washington, D.C.

Barbour, M. T., J. L. Plafkin, B. P. Bradley, C. G. Graves, and R. W. Wisseman. 1992.

Evaluation of the EPA’s rapid bioassessment benthic metrics: metric redundancy and variability among reference stream sites. Env. Tox. and Chem. 11: 437-449.

Burks, B. D. 1953. The mayflies, or Ephemeroptera, of Illinois. Bull. Ill. Nat. Hist. Surv. 26.

216 pp. Davis, W. S., and T. P. Simon. 1995. Biological Assessment and Criteria: Tools for Water

Resource Planning and Decision Making. Lewis Publishers, Boca Raton. 415 pp. DeWalt, R. E., D. W. Webb, and T. N. Kompare. 2001. The Perlesta placida (Hagen) complex

(Plecoptera: Perlidae) in Illinois, new state records, distributions, and an identification key. Proceedings of the Entomological Society of Washington 103: 207-216.

DeWalt, R. E., D. W. Webb, and A. M. Soli. 2002. The Neoperla clymene (Newman) complex

(Plecoptera: Perlidae) in Illinois, new state records, distributions, and an identification key. Proceedings of the Entomological Society of Washington 104: 126-137.

Frison, T. H. 1935. The stoneflies, or Plecoptera, of Illinois. Bull. Ill. Nat. Hist. Surv. 20: 1-471.

CTAP Protocols 37

CTAP Protocols 38

Hilsenhoff, W. L. 1987. An improved biotic index of organic stream pollution. Gr. Lakes Ent. 20: 31-39.

Karr, J. R., K. D. Fausch, P. L. Angermeier, P. R. Yant, I. J. Schlosser. 1986. Assessing

Biological Integrity in Running Waters, A Method and Its Rationale. Illinois Natural History Survey, Special Publication 5: 1-28.

Lenat, D. R. 1993. A biotic index for the southeastern United States: derivation and list of

tolerance values, with criteria for assigning water-quality ratings. J. N. Am. Benth. Soc. 12: 279-290.

Lenat, D. R., and D. L. Penrose. 1996. History of the EPT taxa richness metric. Bull. N. Am.

Benth. Soc. 13: 305-307. Loeb, S. L., and A. Spacie. 1994. Biological Monitoring of Aquatic Systems. Lewis Publishers,

Boca Raton. 381 pp. Rosenberg, D. M., and B. H. Resh, editors. 1993a. Freshwater Biomonitoring and Benthic

Macroinvertebrates. Chapman and hall, New York. 488 pp. Rosenberg, D. M., and B. H. Resh. 1993b. Introduction to freswater biomonitoring and benthic

macroinvertebrates, pp. 1-9 in: D. M Rosenberg, and B. H. Resh, editors. 1993. Freshwater Biomonitoring and Benthic Macroinvertebrates. Chapman and hall, New York. 488 pp.

Ross, H. H. 1944. The caddis flies, or Trichoptera, of Illinois. Bull. Ill. Nat. Hist. Surv. 23: 1-

326. Wallace, J. B., J. W. Grubaugh, and M. R. Whiles. 1996. Biotic indices and stream ecosystem

processes: results from an experimental study. Ecol. Appl. 6: 140-151. Webb, D. W., and M. A. Harris. 1993. Survey of 21 rare species of stoneflies (Plecoptera) in

Illinois. Ill. Nat. Hist. Surv., Tech. Rep. 1993(3): 1-14.

CTAP/INAI Natural Community Categories – Updated, 2005 Forest - Use INAI Categories Prairie - Use INAI Categories. Savanna - Use INAI Categories Wetland - Use INAI Categories CTAP Cultural - See below. Expanded and modified from the INAI categories. This category refers to areas that have been highly disrupted at some point in time by human activities. Generally none, to very little, of the original native vegetation remains.

Miscellaneous categories Developed land – Actively used land that has been highly modified by human activity or contains

structures. Examples include active strip mines, roadways, managed cemeteries, residential and commercial areas, parking lots, buildings, mowed lawns, farm lots, etc.

Utility strip – Linear strips of vegetation, generally planted in non-native grasses and mowed two or three times per year to prevent woody encroachment. Examples include rights-of-way (roadside, railroad, and powerline) and agricultural waterways.

Wet utility strip – Wet utility strips or wet areas in utility strips. Tree plantation – Actively maintained orchards, arboretums, pine plantations, etc. Plantations that

are no longer maintained and are succeeding to native woodlands types are classified as Forests and are acceptable for CTAP monitoring (see CTAP site selection criteria).

Artificial pond – Small to medium sized man-made bodies of water, some of which may have characteristics of natural ponds. CTAP protocols do not include sampling of open water communities, but do incorporate the monitoring of emergent vegetation that often borders these areas.

Artificial lake – Large man-made bodies of water; man-made reservoirs. CTAP protocols do not include sampling of open water communities, but do incorporate the monitoring of emergent vegetation that often borders these areas.

Hayfields - Actively managed agricultural lands generally dominated by planted non-natives such as alfalfa, clover, orchard grass, smooth brome, and timothy. They may also include planted or naturally occurring natives such as switch grass. Indian grass, or bluestem, but they are not prairie restorations (see wildlife planting or prairie restoration). Mowing intensity usually varies from one to three cuts a season. Depending on their age and intensity of seeding they usually have low plant diversity. If these areas have been left unmanaged and woody species are encroaching, they would qualify as old fields.

Hayfield – upland hayfields Wet hayfield – wet hayfields or wet areas in hayfields. These may be seasonally wet and hayed in

dry years. Pastureland - Lands that are actively grazed, or have recently had grazing removed, and are generally

dominated by planted non-native grasses and weedy forbs. Native grasses or forbs may also occur, but the area does not qualify as prairie or prairie restoration. Grazing indicator plants such as gooseberry (Ribes sp.), horsenettle (Solanum sp.), thistle (Cirsium sp.), honey locust (Gleditisia triacanthos), blackberry (Rubus sp.), multi-flora rose (Rosa multiflora), milkweed (Asclepias sp.), St. John's Wort (Hypericum sp.), ironweed (Vernonia sp.), and jimsonweed (Datura sp), and may be common on sites that have undergone heavy grazing. Other signs of grazing may include closely cropped vegetation, livestock trails, bare roots under shade trees, cow pies, and heavy erosion along watercourses. Though crops may have been grown at some point in the land’s history, the area is now obviously maintained as pasture. The site may also be mowed or hayed on occasion. Pastured forestlands are not included in this category.

Pastureland – upland pasture Wet pastureland – wet pasture or wet area in pasture.

Abandoned Lands - These are lands that are not heavily managed, or in obvious agricultural use. They

show obvious signs of succession, or their land history indicates they have been unmanaged for at least 3 years (with the exception of fallow fields). They are usually dominated by adventives. Diversity is often higher than most Hayfields or Pastures, and some shrubs/trees and taller plants will become present as the sites age. Old field – These are former croplands, hayfields, or other agricultural lands, often with a history

of tilled cultivation. They may contain early woody plant succession. Wet old field – These are wet successional fields, or wet areas in successional fields. They may

contain early woody plant succession. Abandoned pasture – These are pastures that are no longer grazed or mowed. Signs of previous

grazing may still be present. Documented land use history confirms past grazing. Wet abandoned pasture – These are wet abandoned pastures, or wet areas in abandoned pastures. Fallow field – This an agricultural field temporarily left fallow that is usually in crop. It is most

likely dominated by weedy annuals, perhaps with some remnants of previous years crops. Wet fallow field – An agricultural area that has recently been left fallow because it is too wet to

farm during the year of sampling, and is characterized by wetland plants. Do not confuse this category with a less transient wetland within a cropped area (i.e. marsh, wet old field). Wet fallow fields are generally dominated by annuals or weedy wetland plants, with evidence of the previous year’s crop stubble. In order to be a wetland, water must be evident or 50% of the dominant species must be FAC to OBL species.

Abandoned utility strip – Utility strips that are no longer maintained or mowed regularly. Abandoned developed land – Developed lands that are no longer actively managed but remain

severely disturbed and have not recovered vegetatively or been planted toward restoration; often with bare soil exposed. Examples include vacant city lots, former industrial areas, former dirt/gravel parking lots, etc.

Restoration - area where the original plant community was removed or destroyed and is now replanted

with conservative species. Diversity can vary depending on the number of species planted. This category may also include CRP lands that are planted in prairie plant mixtures. Prairie restoration – Area planted with the intent to create a native prairie with native grasses and

forbs. These often include acreage enrolled in a conservation programs such as CRP, CREP, etc. Wildflower planting – Areas planted in native species of wildflowers but not with the intent of

restoration to a prairie community. This includes IDOT roadside wildflower/prairies plantings, or wildflower mixes planted by landowners, schools, and other organizations. Though most of these plantings contain many native plants, non-endemics may also be present. These typically include Dame's Rocket, Cosmos, Ox-eye Daisy, etc.

Wetland Restoration – Wet area where the original plant community was removed or destroyed and is now left or managed to recover to a more natural wetland state. There must be intent to restore wetland such as with the wetland reserve program, restoration of the original hydrology (removing tiles, levees), and plantings of wetland species.

Wildlife planting – Areas planted with the intent to encourage use or habitation by wildlife. They are generally planted in native grasses or wildlife food plants and often incorporate acreage included in conservation programs promoted by Natural Resource Conservation Service (NRCS), Illinois Department of Natural Resources (IDNR), Pheasants Forever, etc. They may be mowed or hayed yearly with clippings left on the field.

Woodland restoration – Tree plantings intended to restore an area to natural woodlands. These areas often incorporate acreage included in conservation programs promoted by NRCS, IDNR, or other forestry programs.

Pull down list of Categories Forest

Dry sand forest, Dry upland forest, Dry-mesic sand forest, Dry-mesic upland forest, Forested bog, Forested fen, Mesic floodplain forest, Mesic sand forest, Mesic upland forest, Northern flatwoods, Sand flatwoods, Southern flatwoods, Wet floodplain forest, Wet-mesic floodplain forest, Wet-mesic upland forest, Xeric upland forest

Prairie Dry dolomite prairie, Dry gravel prairie, Dry prairie, Dry sand prairie, Dry-mesic gravel prairie, Dry-mesic prairie, Dry-mesic sand prairie, Glacial drift hill prairie, Gravel hill prairie, Loess hill prairie, Mesic dolomite prairie, Mesic gravel prairie, Mesic prairie, Mesic sand prairie, Sand hill prairie, Shrub prairie, Wet dolomite prairie, Wet prairie, Wet sand prairie, Wet-mesic dolomite prairie, Wet-mesic prairie, Wet-mesic sand prairie

Savanna Dry barren, Dry sand savanna, Dry-mesic barren, Dry-mesic sand savanna, Dry-mesic savanna, Mesic barren, Mesic savanna

Wetland Acid gravel seep, Brackish marsh, Calcareous floating mat, Calcareous seep, Graminoid bog, Graminoid fen, Low shrub bog, Low shrub fen, Marsh, Panne, Sand seep, Sedge meadow, Seep, Shrub swamp, Spring community Swamp, Tall shrub bog, Tall shrub fen

CTAP Cultural Abandoned developed land, Abandoned pasture, Artificial lake, Artificial pond, Developed land, Fallow field, Hayfield, Old field, Pastureland, Prairie restoration, Tree plantation, Utility strip, Wet abandoned pasture, Wet fallow field, Wet hayfield, Wet old field, Wet pastureland, Wet utility strip, Wetland restoration, Wildflower planting, Wildlife planting, Woodland restoration

INAI Communities Forest Upland forest Xeric upland forest Dry upland forest Dry-mesic upland forest Mesic upland forest Wet-mesic upland forest Sand forest Dry sand forest

Dry-mesic sand forest Mesic sand forest

Floodplain Mesic floodplain forest Wet-mesic floodplain forest

Wet floodplain forest Flatwoods Northern flatwoods Southern flatwoods Sand flatwoods Other (see also Wetlands)

Forested bog Forested fen Swamp

Prairie Prairie Dry prairie Dry-mesic prairie Mesic prairie Wet-mesic prairie Wet prairie Sand prairie Dry sand prairie Dry-mesic sand prairie Mesic sand prairie Wet-mesic sand prairie Wet sand prairie Gravel prairie Dry gravel prairie Dry-mesic gravel prairie Mesic gravel prairie Dolomite prairie Dry dolomite prairie Dry-mesic dolomite prairie Mesic dolomite prairie Wet-mesic dolomite prairie Wet dolomite prairie Hill prairie Loess hill prairie Glacial drift hill prairie

Gravel hill prairie

Sand hill prairie Shrub prairie Shrub prairie

Savanna Savanna Dry-mesic savanna

Mesic savanna Sand savanna Dry sand savanna Dry-mesic sand savanna Barren Dry barren

Dry-mesic barren Mesic barren

Wetland Marsh Marsh Brackish marsh Swamp Swamp Shrub swamp Bog Graminoid bog Low shrub bog Tall shrub bog Forested bog Fen Calcareous floating mat

Graminoid fen Low shrub fen Tall shrub fen Forested fen Sedge meadow Sedge meadow Panne Panne Seep & spring Seep Acid gravel seep

Calcareous seep Sand seep Spring community

Lake and Pond Pond Pond Lake Lake Great lake

Stream Creek Low-gradient creek Medium-gradient creek High-gradient creek River Low-gradient river Medium-gradient river Major river CTAP Cultural Miscellaneous Developed land Utility strip Wet utility strip Tree plantation Artificial pond Artificial lake Hayfield Hayfield Wet hayfield Pastureland Pastureland Wet pastureland Abandoned land Old field Wet old field Abandoned pasture Wet abandoned pasture

Fallow field Wet fallow field Abandoned developed land Restoration Prairie restoration

Wildflower planting Wildlife planting Wetland restoration Woodland restoration